The inherited porphyrias are a diverse group of inborn errors of haem biosynthesis, each resulting from the deficient activity of a specific enzyme in the pathway (Fig 1). Porphyrias are classified as either hepatic or erythropoietic, according to whether the excess production of porphyrin precursors and porphyrins occurs primarily in the liver or in the erythron (Anderson et al, 2001). Patients with erythropoietic porphyrias have elevated bone marrow and erythrocyte porphyrins, and their clinical manifestations usually include anaemia and/or cutaneous photosensitivity causing mild to severe dermatological involvement. The erythropoietic porphyrias include congenital erythropoietic porphyria (CEP) and erythropoietic protoporphyria (autosomal dominant ferrochelatase deficiency). The inheritance patterns, enzymatic defects and accumulated porphyrins in the erythropoietic porphyrias are shown in Table I. The human haem biosynthetic pathway. The pathway consists of eight enzymes, four localized in the mitochondria and four in cytoplasm. CEP results from the markedly deficient, but not absent, activity of the fourth enzyme of the haem biosynthetic pathway, uroporphyrinogen III synthase (URO-synthase; EC 4·2.1·75) (Romeo & Levin, 1969; Anderson et al, 2001). In patients with this autosomal recessive disease, the clinical manifestations are markedly heterogeneous, ranging from non-immune hydrops fetalis to milder, later onset forms which have only cutaneous lesions in adult life (Desnick et al, 1998). Severely affected patients are transfusion-dependent throughout life, have secondary hypersplenism and marked cutaneous involvement. Here, we present the clinical features, molecular genetics and pathology, and the successful treatment by bone marrow transplantation (BMT) of patients with this erythropoietic porphyria. CEP was first described by Schultz (1874) in a 33-year-old weaver who had cutaneous photosensitivity since the age of 3 months, splenomegaly, anaemia and wine-red urine (porphyrinuria). This was the first patient reported with a porphyria. The patient's urine was studied by Baumstark (1874) who identified two pigments which he named urorubrohaematin and urofuscohaematin. He noted that an acid solution of urorubrohaematin was similar to the acid solution of haematoporphyrin described by Hoppe-Seyler (1871) and suggested that these pigments resulted from an error in the biosynthesis of haemoglobin. Additional early cases, recognized by the unusual porphyrinuria, further delineated the clinical features of the disease (Gagey, 1896; McCall-Anderson, 1898; Vollmer, 1903). Günther (1911) defined this entity as 'haematoporphyria congenita' and was the first to recognize it as an inborn error of metabolism, and provided Hans Fischer with the early patient samples that stimulated his studies of pyrrole chemistry. In fact, Fischer had the edge over other chemists competing in the structural characterization of the porphyrins as he recruited Mathias Petry, a famous patient with CEP, to his laboratory as a technician and source of urine and stool specimens. When Petry died in 1925, Fischer performed an extensive biochemical and pathologic autopsy which identified the erythroid, rather than hepatic, pathophysiology of the disease (Fisher et al, 1925). In the mid 1950s, Schmid and co-workers further characterized the erythroid nature of the disease and suggested that this congenital disorder be more appropriately named 'porphyria erythropoietica' (Schmid et al, 1954, 1955). Subsequently, the disease became known as congenital erythropoietic porphyria, which clearly distinguished this recessive disorder from erythropoietic protoporphyria, a disease which typically has a clinical onset later in childhood. CEP also has been termed Günther disease, erythropoietic porphyria, congenital porphyria, congenital haematoporphyria and erythropoietic uroporphyria. Although the chemical nature of the accumulating porphyrins in patients with CEP was delineated by 1930 (Moore, 1990), it was not until 1969 that the specific enzyme defect in CEP was identified (Romeo & Levin, 1969). The finding of markedly deficient URO-synthase activity in the cells of patients with CEP closed an almost 100-year period from the first description of the disease and subsequent characterization of the accumulated substrates to the identification of the specific enzyme defect causing this inborn error of haem biosynthesis. Recent studies have focused on the isolation and characterization of the URO-synthase enzyme and gene, the nature of the genetic lesions causing CEP and effective treatment by bone marrow transplantation. CEP is rare and panethnic. Patients have been identified in such diverse racial and demographic groups as Japanese, Northern Europeans, Indians, Hispanics, African Americans and Cree Indians. As of 2000, only about 150 cases had been reported in the literature (Fritsch et al, 1997; Desnick et al, 1998). CEP also occurs in animals such as cattle (Moore et al, 1987) and the fox squirrel (Levin & Flyger, 1971; Levin & Flyger, 1973; Flyger & Levin, 1977). The age at onset and clinical severity of CEP are highly variable, ranging from non-immune hydrops fetalis due to severe haemolytic anaemia in utero to milder, later-onset forms, which have only cutaneous lesions in adult life. In most cases, photosensitivity develops soon after birth. A number of factors are responsible for the phenotypic variability including: (1) the amount of residual URO-synthase activity (Warner et al, 1992a), (2) the degree of haemolysis and consequent stimulation of erythropoiesis, and (3) exposure to ultraviolet light. The major clinical features are listed in Table II, including the markedly debilitating anaemia and cutaneous photosensitivity. Overall life expectancy may be markedly diminished in more severely affected patients due to haematological complications and the increased risk of infection. Mild to severe haemolysis is accompanied by anisocytosis, poikilocytosis, polychromasia, basophilic stippling, reticulocytosis, increased nucleated red cells, absence of haptoglobin, increased unconjugated bilirubin and increased faecal urobilinogen. Plasma iron turnover also is increased (Schmid et al, 1955). Howell–Jolly bodies are often present and the bone marrow contains 30–70% of normoblasts that have porphyrin florescence. Haemolysis presumably results from the accumulated uroporphyrin I in erythrocytes. Secondary splenomegaly develops in response to the increased uptake of abnormal erythrocytes from the circulation, which may contribute to the anaemia and also may result in leucopenia and thrombocytopenia. The latter is sometimes associated with significant bleeding and splenectomy may be beneficial in such cases (Pain et al, 1975; Weston et al, 1978). Anaemia due to haemolysis can be severe. Erythroid hyperplasia and markedly ineffective erythropoiesis usually accompany haemolytic anaemia in transfusion-dependent patients. In most cases of CEP, cutaneous photosensitivity usually begins in early infancy and is manifested by increased friability and blistering of the epidermis on the hands and face and other sun-exposed areas. Bullae and vesicles contain serous fluid which fluoresces owing to their porphyrin content; they are prone to rupture and infection. The skin may be thickened, with areas of hypo- and hyper-pigmentation. Hypertrichosis of the face and extremities is often prominent (Poh-Fitzpatrick, 1986). Sunlight, other sources of ultraviolet light and minor skin trauma increase the severity of the cutaneous manifestations. Recurrent vesicles and secondary infection can lead to cutaneous scarring and deformities, as well as to loss of digits and facial features such as the eyelids, nose and ears (Fig 2). Adult-onset patients have milder clinical symptoms and often exhibit only the skin manifestations of the disease (Kramer et al, 1965; Pain et al, 1975; Deybach et al, 1981; Murphy et al, 1995). Patients with CEP showing destructive cutaneous manifestations. (A) Face and (B) hands of a 14-year-old male (genotype C73R/T228M); (C) hands of a 37-year-old woman (genotype G225S/T228M). Corneal scarring can lead to blindness. Porphyrins deposited in the teeth produce a reddish-brown colour in natural light, termed erythrodontia. The teeth may fluoresce on exposure to long-wavelength ultraviolet light. Porphyrin deposition in bone also occurs, and bone loss due to demineralization (Piomelli et al, 1986) or expansion of the hyperplastic bone marrow (Laorr & Greenspan, 1994; Anderson et al, 2001) has been described. The pathogenesis of CEP is readily explained by markedly deficient, but not absent, URO-synthase activity. Consistent with its autosomal recessive inheritance, a very low level (≤1% of normal) appears to be sufficient for life (see 'Genotype/Phenotype Correlations'). The deficient URO-synthase activity leads to the accumulation of the substrate hydroxymethylbilane (HMB), most of which is converted non-enzymatically to uroporphyrinogen I (Fig 3). Although uroporphyrinogen I can undergo decarboxylation by uroporphyrinogen decarboxylase to form hepta-, hexa- and pentacarboxyl porphyrinogen I and, finally, coproporphyrinogen I, further metabolism cannot proceed because the next enzyme in the pathway, coproporphyrinogen oxidase, is stereospecific for the III isomer. Therefore, the isomer I porphyrins are non-physiological, in that they cannot be metabolized to haem, and are pathogenic when they accumulate in large amounts and are auto-oxidized to their corresponding porphyrins. In patients with CEP, the large amounts of isomer I porphyrinogens that accumulate in bone marrow erythroid precursors (especially normoblasts and reticulocytes) and erythrocytes undergo auto-oxidation to the corresponding porphyrins, which damage erythrocytes, cause cutaneous photosensitivity, are deposited in tissues and bones, and are excreted in large amounts in the urine and faeces. Photosensitivity occurs because these porphyrins are photocatalytic and cytotoxic compounds (Poh-Fitzpatrick, 1985). Exposure of the skin to sunlight and other sources of long-wave ultraviolet light results in blistering and vesicle formation and increased friability of the skin (Bickers, 1987) (Fig 2). Haemolysis is almost always present, but may not be accompanied by anaemia if erythroid hyperplasia is sufficient to compensate for the increased rate of erythrocyte destruction. The degree of compensation may vary over time. The enzyme defective in CEP. Hydroxymethylbilane is normally converted to uroporphyrinogen III by URO-synthase. The defective activity of this enzyme results in the non-enzymatic conversion of hydroxymethylbilane to the uroporphyrinogen I isomer. Although uroporphyrinogen I can undergo decarboxylation by uroporphyrinogen decarboxylase to form hepta-, hexa- and pentacarboxyl porphyrinogen I and, finally, coproporphyrinogen I, further metabolism cannot proceed because the next enzyme in the pathway, coproporphyrinogen oxidase, is stereospecific for the III isomer. Uroporphyrin I accumulation in bone marrow, erythrocytes, plasma and urine is the biochemical hallmark of the disease. Pink to dark-reddish urine is usually noted shortly after birth. Urinary porphyrin excretion is markedly increased (100–1000 times normal; up to 50–100 mg/d) and consists mostly of uroporphyrin and coproporphyrin, with lesser increases in hepta-, hexa- and pentacarboxyl porphyrins (Fritsch et al, 1997). Although the predominant increase is in the type I isomers, type III isomers also are increased. Urinary aminolevulinic acid (ALA) and porphobilinogen (PBG) excretion are not increased. Faecal porphyrins are markedly increased with a predominance of coproporphyrin I. Faecal isocoproporphyrins are not increased (Verstraeten et al, 1993). Circulating erythrocytes in most reported cases of CEP have contained large amounts of uroporphyrin I and lesser, but still excessive, amounts of coproporphyrin I. Red cell protoporphyrin also may be increased and in some reported cases was the predominant porphyrin in erythrocytes, as in the bovine form of this disease (Watson et al, 1959; Moore et al, 1987). The bone marrow contains much larger amounts of porphyrins than other tissues in this disease, and these consist mostly of uroporphyrin and coproporphyrin. These porphyrins are also found in spleen and to a lesser extent in liver (Schmid et al, 1954, 1955; Rosenthal et al, 1955; Schwartz et al, 1976). Excess plasma porphyrins, which probably originate from the bone marrow and circulating red cells, are mostly uroporphyrin and coproporphyrin. Erythrocyte HMB-synthase (also known as porphobilinogen deaminase) is increased in this disease and is attributable to increased erythropoiesis (Romeo et al, 1970), as in other disorders associated with haemolysis (Anderson et al, 1977). Studies of nine Indian CEP patients indicated a close relationship between the degree of porphyrin excess and the severity of disease expression (Freesemann et al, 1997). URO-synthase normally catalyses the conversion of the linear tetrapyrrole hydroxymethylbilane by inversion of the pyrrole D ring and cyclization to uroporphyrinogen III, the physiological cyclic isomer which is metabolized in subsequent enzymatic steps to haem (Fig 1) (Jordan & Berry, 1980; Battersby et al, 1982a,b; Shoolingin-Jordan, 1995). In patients with CEP, URO-synthase activity is markedly deficient, but not absent. Presumably, the total absence of URO-synthase activity would be 'fetal lethal' as it is in the homozygous null mouse (Bensidhoum et al, 1998). Even severely affected homozygotes who present with hydrops fetalis must have sufficient enzymatic activity to synthesize adequate amounts of uroporphyrinogen III and haem for viability. The markedly deficient URO-synthase activity results in the non-enzymatic conversion of hydroxymethylbilane to the uroporphyrinogen I isomer (Fig 3). As noted above, this non-physiological and pathogenic compound can be metabolized to coproporphyrinogen I, but further metabolism cannot proceed as the next enzyme in the pathway, coproporphyrinogen oxidase, is stereospecific for the III isomer. The accumulated uroporphyrinogen I and coproporphyrinogen I are then oxidized to the non-physiological and presumably pathogenic porphyrins, uroporphyrin I and coproporphyrin I, respectively. URO-synthase has been purified to homogeneity from human erythrocytes and shown to be a monomeric protein with an apparent molecular weight of 29·5 kDa (Tsai et al, 1987a). The purified enzyme had a specific activity of over 300 000 nmol/h/mg, an isoelectric point of 5·5 and was remarkably thermolabile (t1/2 at 60°C ∼1 min). The enzyme's pH optimum was 7·4 and the Km for HMB was 5–20 mmol/l. The enzyme was activated by Na+, K+, Mg2+ and Ca2+, and was inhibited by Cd2+, Cu2+, Hg2+ and Zn2+. In human erythrocytes, there is an excess of URO-synthase activity compared with HMB-synthase, thus favouring the synthesis of uroporphyrinogen III over uroporphyrinogen I. It has been suggested that URO-synthase and HMB-synthase exist in a cytosolic complex (Tsai et al, 1987a). Sensitive assays have been developed to measure URO-synthase activity in erythrocytes and cultured cells (Tsai et al, 1987b). URO-synthase also has been purified from spinach (Higuchi & Bogorad, 1975), Euglena gracilis (Hart & Battersby, 1985), Escherichia coli (Alwan et al, 1989) and rat liver (Smythe & Williams, 1988). The cDNA encoding human URO-synthase was isolated, sequenced and expressed in E. coli (Tsai et al, 1988). The originally isolated cDNA was 1296 bp with 5′ and 3′ untranslated regions of 196 bp and 302 bp, respectively, and an open reading frame of 798 bp that encoded a polypeptide of 265 amino acids. Using the cDNA as a probe, a single URO-synthase gene was assigned to the narrow chromosomal region 10q25.3→q26.3 (Astrin et al, 1991). The genomic structure of the human URO-synthase gene recently was determined (Aizencang et al, 2000). The gene is ∼34 kb in length and contains 10 exons, ranging from 75 to 423 bp in length, while the introns are ∼0·2–8·9 kb (Fig 4). All intron–exon boundaries conform to the GT-AG rule. Of particular interest, the URO-synthase gene has alternative promoters that generate housekeeping and erythroid-specific transcripts. The housekeeping transcript contains the 5′ untranslated exon 1 fused to coding exons 2B through 10 and is present in all tissues and cells, while the erythroid-specific transcript contains the alternative 5′-untranslated exon 2A fused to coding exons 2B through 10 and is present in fetal and adult erythropoietic tissues. The housekeeping proximal promoter region upstream of exon 1 lacked TATA and SP1 sites for initiation of transcriptional activity of the housekeeping transcript, consistent with its observed low level of expression in most cells. The erythroid-specific proximal promoter upstream of exon 2A contained erythroid transcription factor binding sites including GATA1 and NF-E2. Schematic diagram of the human URO-synthase gene. The housekeeping and erythroid transcripts are shown below the gene. ATG and TGA indicate the translation start and stop sites respectively. The arrows indicate the housekeeping (HPr) and erythroid-specific (EPr) promoter regions. Luciferase reporter assays performed in erythroid K562 and HeLa cells demonstrated that the housekeeping promoter was active in both cell lines, while the erythroid-specific promoter was active only in erythroid cells. The relative expression levels of the erythroid-specific and housekeeping transcripts were surveyed by hybridization with an RNA array of various human tissues. High levels of erythroid expression (exon 2A containing mRNAs) were detected in fetal liver, bone marrow and fetal spleen, while only background levels were observed in other tissues. In contrast, hybridization of the RNA with the housekeeping probe revealed the ubiquitous expression of the exon 1-containing transcripts, the highest being in the caudate nucleus and amygdala (Fig 5) (Aizencang et al, 2000). Thus, the genomic organization of the URO-synthase gene was similar to those of the second and third genes in the haem biosynthetic pathway, 5′-aminolevulinic (ALA)-dehydratase and HMB-synthase, in having alternative housekeeping and erythroid-specific promoters. Of note, the first four genes in the haem biosynthetic pathway have unique genomic structures for erythroid-specific expression (Fig 6) (Grandchamp et al, 1987; Bishop, 1990; Kaya et al, 1994; Aizencang et al, 2000). Expression of the housekeeping and erythroid URO-synthase (UROS) mRNA transcripts. A Multiple Tissue Expression array containing equalized concentrations of poly(A)+ RNAs from different human tissues was hybridized with specific probes for the (A) housekeeping and (B) erythroid URO-synthase transcripts and (C) tryrosine (Tyr) aminotransferase transcripts. Gene expression results were calculated as the percentage of maximum tissue levels. The inset in (B) shows the Northern blot of erythroid tissue mRNA hybridized with the URO-synthase exon 2A probe. Genomic structures of the first four enzymes of the haem biosynthetic pathway showing the transcripts formed from two different ALA-synthase genes or from the alternative housekeeping and erythroid-specific promoters of ALA-dehydratase, HMB-synthase and URO-synthase. The mouse cDNA and genomic sequences encoding URO-synthase also have been isolated and characterized (Bensidhoum et al, 1994; Xu et al, 1995a; Aizencang et al, 2000). The mouse cDNA had 5′- and 3′-untranslated regions of 144 and 623 bp, respectively, and an open reading frame which also encoded 265 amino acids. Comparison of the human and mouse cDNAs and the predicted amino acid sequences revealed 80·5% nucleotide and 77·8% amino acid identity (Xu et al, 1995a). The murine gene was localized to a narrow region on mouse chromosome 7 within 1·5 cM of Oat, which is syntenic with the localization of the human gene on chromosome 10q25.3→q26.3 (Astrin et al, 1991). Like the human URO-synthase gene, the mouse gene has distinct housekeeping and erythroid-specific promoters. Availability of the mouse gene should permit the development of mouse knockout and knock-in models for CEP. Recently, the three-dimensional structure of recombinant human URO-synthase was determined by X-crystallography at 1·85 Å resolution (Mathews et al, 2001). The monomeric protein was folded into a two-domain structure, each domain having a parallel β-sheet surrounded by α-helices. The two domains were connected by a two-strand, anti-parallel β-ladder (Fig 7). Presumably, the active site is in the large open cleft between the two domains, where surface-exposed conserved residues are clustered. It is hypothesized that the intradomain cleft is flexible and that its more open conformation may facilitate substrate binding and product release, while its more closed conformation may be involved in ring closure. Model of URO-synthase crystal structure bound to uroporphyrinogen III and showing the location of 16 point from patients with CEP. et by of The of the URO-synthase cDNA and genomic sequences has the identification of the molecular lesions causing CEP. that cause CEP have been reported in the URO-synthase gene et al, Xu et al, Desnick et al, Anderson et al, et al, et al, and the Gene These include and large and and and erythroid-specific promoter Of the and four and at known for et al, & Of interest, the most URO-synthase not at a All but of the known CEP in amino acid residues that were also conserved in the murine URO-synthase polypeptide (Bensidhoum et al, 1994; Xu et al, 1995a). in exon 10 has been identified in CEP patients. in the URO-synthase gene causing CEP. are shown in Expression of the in E. coli indicated the amount of residual activity encoded by these In this expression encoded enzymes with low of the expressed if levels of URO-synthase activity These and et al, Desnick et al, 1998). expressed levels of enzymatic of the expressed and and had levels of residual ranging from of that expressed by the they were than the expressed having at of and (Desnick et al, et al, Of interest, the and lesions also as described These were associated with milder (see of URO-synthase on the recently crystal structure of the enzyme suggested that most the enzyme's structure, resulting in or (Fig et al, 2001). two of the and in residues that were the active site It is not that the of the were that not markedly the enzyme's structure as most of the enzymes very enzymatic consistent with the that affected not in the and were of particular as each was a single in the to or to nucleotide of exon and the exon 5′ respectively. had significant residual URO-synthase activity when expressed in E. and of expressed respectively, but also defects (Xu et al, et al, transcription of the transcripts encoded by the and revealed and exon Thus, these two with residual activity had only a of activity due to and would be predicted to cause a to severe on the other the had only of transcripts which expressed of resulting in than of activity. In to the that in the 5′ or 3′ site sequences These a to of the 5′ site of the of an in the of the 5′ site of (Xu et al, and a in the of et al, the URO-synthase gene have two and et al, et al, and et al, et al, Xu et al, et al, and complex et al, gene or have been The of alternative housekeeping and erythroid-specific promoters in the human and murine URO-synthase genes the identification of promoter point within a region of the erythroid-specific promoter have been identified in CEP et al, 2001). These a to a binding (2) a to A (3) a to A in and a to A which a binding of with the or into K562 erythroid cells resulted in 3 and of activity respectively. assays indicated that the GATA1 the the the Thus, these four pathogenic erythroid promoter erythroid-specific CEP, and identified GATA1 and transcriptional binding for erythroid-specific haem biosynthesis. these are the only known promoter in the erythropoietic promoter has been identified in a hepatic porphyria et al, 2000). The most was found in a third of the CEP studied of CEP the URO-synthase gene that the of was due to et al, 1998). These indicate that is a and that the of not result from a The next most were and and Table for the the other have been detected in only or a CEP and, were of the were in having been identified in diverse racial and demographic expression and gene have been to the in levels of enzymatic activity expressed by the and the promoter of in the erythroid-specific As noted above, the expression of URO-synthase containing resulted in levels of enzymatic activity that from to about of the activity expressed by the in E. coli The of the four promoter on transcription also was in by the activity of each gene in and K562 cells. a of CEP patients were classified as very mild to severely affected on degree of haemolytic and cutaneous for the most was with the most severe non-immune hydrops fetalis and/or from birth. Consistent with the severe of expression of the in E. coli resulted in the of of the activity expressed by the The that the
No takes yet. Share an insight, caveat, or question.
Dsnick et al. (2002) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: